 I'm very glad to take part in the Cambridge International Law Conference, even if it's a little bit disappointing to talk from my office in my house here in Yesh, Belgium. We live in a risk society and we feel it more than ever with the COVID-19 crisis. Risk and law are interconnected concepts in many ways. I will focus on the concept of risk as it is used in the case law of the European Court of Human Rights. My presentation will consist of three parts. You can have a look at the summary. Like many other institutions, the European Court of Human Rights is often called upon to deal with complex situations where the notion of risk is at stake. The most common issues include suicides of individuals who were under state's control like in prison or military service, but also violence against individuals, damages to persons or properties caused by natural disasters or industrial accidents. In these cases, applicants often claim that they are today or have been in the past, exposed to a particular risk and alleged that a state fails or had failed to react adequately in order to prevent a damage that can be death or injury. In these situations, the analysis that the judges have to produce is a very complex intellectual challenge. They have to assess a current or a past risk and deal with the uncertainty, with the potentiality of a damage induced by a fact or a decision. Judges are like time travelers without any time machine. So how do they do to deal with risks and decide whether states respond to them in a manner consistent with human rights? I don't have a final answer to this question, but some elements that can be considered. I've tried to analyze in details the terminology used by the court in the judgments in which the notion of risk is a decisive element for the solution of the case. And I'd like to show you what we can learn from this terminology about the kind of risk that can be the basis of a human right violation. As the time is limited, I will talk about three findings, the first of which is the diversity of adjectives used by the court to qualify the risks it is considering. To this difficulty should be added a second one. The use of words by the court is not always perfectly consistent. This is particularly striking when comparing the English and French versions of some judgments. I take a very simple example from the Roman Vibagian judgment that I don't want to analyze in details here. The point is that the word risk is used twice in the English version, as you can see on this slide, but is translated by two different words in French, risk and danger, which obviously corresponds to the word danger in English. I know that these kinds of inconsistencies can be found in other areas of the case law, but it perhaps shows that the concepts relating to risk are not really stabilized. But the third finding is maybe the most important. Among the relevant formulas used by the court, some are more frequent than others. In many cases, what is decisive and may under certain conditions indicate the existence of a violation of a human right is the identification of a real risk, or a real and immediate risk. In this regard, the typical formula in the case law is the following one. For a positive obligation to arise, it must be established that the authority is new or out to have known of the existence of a real and immediate risk. It could be a risk, for example of ill treatment in the case of children. But what is a real risk? According to the Oxford Fictionary online, something real is actually occurring in fact. It is not imagined or supposed. It seems to mean that there should be no doubt about the existence of a risk. What does it mean? What could we conclude? In my opinion, the words used by the court, the terminology used by the court, do not express any kind of analysis of the level of the risk. But the search for evidence of a certain risk. Is there enough evidence to prove the existence of a certain risk? Even in cases where the notion of risk is at stake, even if these situations are characterized by uncertainty, the court as a legal body is looking for evidence. Not the evidence of a damage, but the evidence of a risk. This analysis of the terminology used by the court is one thing. But there is a need for further reflection. This is why in the last part of my presentation, I'd like to extend the scope to the overall reasoning of the court. In the cases where the notion of risk is at stake, I think that it is possible to find references to the key elements of the risk literature, especially of the economic risk literature. In this regard, the common risk analysis is basically founded under the determination of two elements. The level of a risk is the product of the values of these factors. The likelihood or the probability of a damage and the severity of this damage. To do this kind of assessment, risk managers rely on matrices such as this one. When the level of a risk has been assessed, the next step is to examine the acceptance of this risk. A low risk is most of the time more acceptable than a high risk. But it is not always the case because many parameters need to be taken into account. The main idea can be understood with a simple question. Is the benefit expected when taking the risk higher than the level of the risk itself? In practice, the answer to this question depends on cultural and psychological parameters. The subjective dimension is very significant. So you have three main concepts, severity, likelihood and acceptance. In my opinion, it's arguable that all of them play a role explicitly or implicitly when compliance to human rights is discussed in combination with the notion of risk. In many cases, to co-opt exomines, whether the damage or the potential damage exceeds a minimum level of severity. When a sufficiently serious potential damage is identified, the other parameters, likelihood and acceptance, can be used to decide whether a violation of the convention is finally found. The assessment of the likelihood of damage is sent probably in many cases involving risks. The court refers to various data provided by the parties or collected by itself. I take one example. When the court has to assess the risk of built treatment in the event of the expulsion of a person to another country, it relies on reports produced by governmental or non-governmental sources, which describe the situation in that country regarding the compliance with human rights. It helps to estimate the probability of damage. Two minutes ago, I've shown the typical matrix that is used by risk managers. The matrix of the court is a binary one. When it combines its analysis of the severity and of the likelihood of damage, there are only two possible results. Either there is no relevant risk or there is a real and immediate risk. The last concept to discuss is the idea of risk acceptance. As I said a moment ago, whether a risk is acceptable or not is a matter of subjective judgment. Is there an assessment of the acceptance of risks in the case law of the European Court of Human Rights? Certainly not explicitly. But I think that the notion of risk acceptance can provide explanations for legal concepts that also have an elastic character, such as legitimacy, proportionality, or margin of appreciation. I only have the time for two kinds of considerations. Firstly, some risks are acceptable because the states have a limited material capacity. You can think about financial capacity or human resources. According to the court, the positive obligation does not be interpreted in such a way as to impose an impossible or disproportionate burden on the authorities. Some risks are also acceptable because the states have a limited legal capacity. Even when they have the material resources, they cannot do what they want. We come to the end of this presentation. I insist that my study is a work in progress and I will welcome your comments to further improve it. I thank you very much for your attention.